The custom designed module packages Google’s Edge TPU ASIC within a miniaturised footprint. The solution looks to overcome some of the challenges associated with implementing AI solutions by delivering improved noise suppression and simplifying printed circuit board design in a smaller footprint.
Miniaturisation is seen as key, as all board space must be optimised to achieve highly robust functionality in space constrained operations. The result of this collaboration is a solution that is said to speed up the algorithmic calculations required to execute AI.
“Coral enables new applications of on-device AI across many industries, from manufacturing to healthcare to agriculture. Working with Murata to make the Coral Accelerator Module – with Google Edge TPU – available in a robust, solderable, and easy-to-integrate package means that more customers can include Coral intelligence inside their products in more environments,” said Vikram Tank, Product Manager for Coral.
The goal of Coral is to enable AI applications running at the device level to quickly move from prototype to production. Coral provides the complete toolkit of hardware components, software tools, and pre-compiled models for building devices with local AI. The AI module is an integral part of the fully integrated Coral platform, which can be implemented in a myriad of applications across numerous industries.
Murata worked closely with Coral to ensure that the AI module helped enable the flexibility, scalability, and compatibility for integration into applications deploying the Coral technology. Toward this end, Murata leveraged its global resources and decades of R&D in the areas of high-density design and component integration.
